INDEX

Explanations

prefixes and word beginnings

New Auto-Interp

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Problemas

-1.30

 Evaluación

-1.22

 Según

-1.22

mitteilung

-1.21

<bos>

-1.21

 Pueden

-1.18

 Aún

-1.17

 Ofrece

-1.16

 ภาพ

-1.15

大根

-1.14

POSITIVE LOGITS

not

1.47

 each

1.38

 every

1.37

it

1.27

 your

1.27

its

1.27

 their

1.25

all

1.24

of

1.16

 them

1.15

Activations Density 0.220%